#1094 Remove the ambient ripgrep dependency from the test suite - #1097
Merged
Conversation
Dependency auditProduction dependency audit passed. |
7 tasks
Running the recall tests no longer requires ripgrep on PATH. Recall now fails loudly when ripgrep reports matches whose output it cannot parse, rather than returning the same empty result a vault with nothing to find would produce.
Running the search tests no longer requires ripgrep on PATH. A test names the notes recall found instead of relying on a query matching the right fixture notes, so what it asserts about scoping, filtering, and projection no longer depends on the matcher.
Running the retrieve CLI tests no longer requires ripgrep on PATH. Each test states the notes recall found, so what it asserts about candidate tables, filters, and diagnostics no longer depends on a query happening to match the right fixture notes. The --diataxis test recalls a reference note alongside the how-to notes, giving the filter it exercises something to exclude.
The post-build smoke run no longer requires ripgrep on PATH, so no step of the build depends on an external binary. The kb-retrieve-events pairing scopes to an unregistered store, asserting the resolver reached a scope verdict and shaped its JSON result, rather than recalling a seeded event.
The recall stub builds its missing-path set from the optional argument directly, and the search test names a nested fixture path without composing two joins.
Comments on the recall seam, its stub, and the smoke pairing state the contract and the non-obvious invariants alone, dropping the rationale for introducing them and the description of what the tests did beforehand.
`parseRipgrepOutput` returns at most one entry per note path, so `searchKb` builds its hits with a direct map rather than a keyed map whose duplicate check can never fire.
Each of these tests declares the notes recall found, so the surviving candidate count is exactly one. The assertions state that count rather than a lower bound.
`multi-match.md`, `2026-05-01-meeting-notes.md`, `2024-archive/legacy-runbook.md`, and `2026-06-01/daily-log.md` each carried a unique term for a recall test that matched it through the real binary. Those tests now assert the ripgrep argument list and parse synthetic events, so no test reads any of the four notes.
williamthorsen
marked this pull request as ready for review
July 28, 2026 18:17
williamthorsen
added a commit
that referenced
this pull request
Aug 4, 2026
…e-v0.2.2 codeassembly-v0.4.0 factory-v0.2.2 codeassembly-mcp-v0.2.2 codeassembly-v0.4.0 - #1153 feat: Make codeassembly and kb CLI tools publishable (#1164) - #1091 fix: Anchor a project-deployed link where its target deploys (#1159) - #1156 refactor: Rename packages to publishable names (#1157) - #1152 tooling: Run every test in the default gate, classified by what it reaches (#1155) - #1110 tooling: Migrate Vitest to nmr's centralized model (#1154) - refactor: Refine typescript-preferences - #1150 fix: State doc-description form and make comment mood opt-in (#1151) - deps: Upgrade first-party linting deps to latest version - #1137 feat: Establish personal rulebooks for code layout and TypeScript preferences (#1144) - #1126 feat: Check a package's own guidance content before it ships (#1138) - refactor: Fix lint - tooling: Remove redundant lint rules - #1133 fix: Reject an anchor link that names no heading (#1135) - #1122 feat: Capture lede decisions as an accumulating corpus (#1132) - #1125 feat: Honor invocation tokens in rulebook bodies (#1129) - #1107 feat: Render rulebook links and path tokens per harness (#1124) - #1115 feat: Sync guidance at build and install so an upgrade cannot leave it stale (#1123) - #1114 feat: Adopt a dependency's guidance by naming the package (#1121) - #1088 feat: Deliver project ambient rulebooks per harness (#1113) - #1087 refactor: Rename the authoring rulebook and mark its enforced rules (#1092) - #1095 tooling: Move compilation out of the install lifecycle into a bootstrap step (#1102) - #1094 refactor: Remove the ambient ripgrep dependency from the test suite (#1097) - deps: Upgrade all deps to latest version - tooling: Upgrade all deps to latest version & modernize configs - refactor: Fix lint - #1077 feat: Show the proposed edit above post-review menu options (#1078) - #1068 feat: Deliver ambient rulebooks mechanically, retiring GLOBAL.md (#1075) - #1069 feat: Add a no-second-person rule to the lede-voice doctrine (#1070) - #1050 feat: Add a redundancy rule to the lede-voice doctrine (#1058) - #1035 internal: Add lifecycle workspace with the canonical envelope, vocabulary & lane fold (#1049) - #1006 internal: Extend lifecycle-event instrumentation to five high-traffic skills (#1034) - #1028 feat: Rule out absence-of-removed-code tests (#1032) - #114 feat: Let reviewers emit gated insights into review artifacts (#1031) - #1027 internal: Retire input.received and redundant skill.progress emits (#1030) - #1026 fix: Name the side effects an approval ask authorizes (#1029) - #1022 fix: Normalize action and question label identifiers across asks blocks (#1025) factory-v0.2.2 - #1156 refactor: Rename packages to publishable names (#1157) - deps: Upgrade all deps to latest version - #1152 tooling: Run every test in the default gate, classified by what it reaches (#1155) - #1110 tooling: Migrate Vitest to nmr's centralized model (#1154) - tooling: Remove redundant lint rules - deps: Upgrade deps to latest version - deps: Upgrade all deps to latest version - deps: Upgrade all deps to latest version - tooling: Upgrade all deps to latest version & modernize configs - refactor: Fix lint - deps: Upgrade all deps to latest version kb-v0.3.0 - #1153 feat: Make codeassembly and kb CLI tools publishable (#1164) - #1156 refactor: Rename packages to publishable names (#1157) - #1152 tooling: Run every test in the default gate, classified by what it reaches (#1155) - #1110 tooling: Migrate Vitest to nmr's centralized model (#1154) - tooling: Remove redundant lint rules - #1095 tooling: Move compilation out of the install lifecycle into a bootstrap step (#1102) - tooling: Upgrade all deps to latest version & modernize configs - refactor: Fix lint codeassembly-lifecycle-v0.2.0 - #1153 feat: Make codeassembly and kb CLI tools publishable (#1164) - #1156 refactor: Rename packages to publishable names (#1157) - #1152 tooling: Run every test in the default gate, classified by what it reaches (#1155) - #1110 tooling: Migrate Vitest to nmr's centralized model (#1154) - tooling: Remove redundant lint rules - #1095 tooling: Move compilation out of the install lifecycle into a bootstrap step (#1102) - tooling: Upgrade all deps to latest version & modernize configs - #1038 feat: Add the read-only git adapter for worktree and base-branch ground truth (#1059) - #1051 feat: Bound fold memory and rescan cost with a retention window (#1057) - #1035 internal: Add lifecycle workspace with the canonical envelope, vocabulary & lane fold (#1049) codeassembly-mcp-v0.2.2 - deps: Upgrade all deps to latest version - #1156 refactor: Rename packages to publishable names (#1157) - #1152 tooling: Run every test in the default gate, classified by what it reaches (#1155) - #1110 tooling: Migrate Vitest to nmr's centralized model (#1154) - #1095 tooling: Move compilation out of the install lifecycle into a bootstrap step (#1102) - tooling: Upgrade all deps to latest version & modernize configs - refactor: Fix lint codeassembly-run-core-v0.2.2 - #1156 refactor: Rename packages to publishable names (#1157) - #1152 tooling: Run every test in the default gate, classified by what it reaches (#1155) - #1110 tooling: Migrate Vitest to nmr's centralized model (#1154) - #1095 tooling: Move compilation out of the install lifecycle into a bootstrap step (#1102) - tooling: Upgrade all deps to latest version & modernize configs - refactor: Fix lint
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
What
Tests and the post-build smoke check no longer depend on ripgrep. Separately, the knowledge-base retrieval skills now fail with an explicit error when ripgrep returns output they cannot interpret, instead of reporting an empty result when matching notes exist.
Why
Thirty-seven tests across four files required
rgonPATH, and a workflow modernization that dropped the step installing it left the build red. Most of those tests had no stake in ripgrep, reaching it only in passing, so an absent binary failed tests about candidate tables and filters with an error naming the recall module rather than the missing prerequisite.Taking every test off the real binary also removes the only thing that would have caught a change to ripgrep's
--jsonevent format, so recall has to detect that condition itself.Details
♻️ Refactoring
recallNotesaccepts an optional process runner and reachesrgonly through it, so its own tests assert the constructed argument list and exercise the exit-code and missing-binary branches without spawning anything. The runner's contract matches the shapepackages/fleet/src/forge/github-adapter.tsalready establishes.searchNotesaccepts an optional recall function, threaded throughrunRetrievein both retrieve CLIs, so a test of scoping, filtering, or projection supplies hits directly and never reaches the process layer. The seam is recall rather than the process runner because threading the runner that far would force consumer tests to author ripgrep--jsonevents carrying machine-absolute paths, which is matching reimplemented in test code.searchKbthrows when ripgrep returns output that parses to no entries. ripgrep exits 1 when nothing matched, so a zero exit carrying output guarantees at least one match event, and parsing none of it means the--jsonevent shape no longer matches what the module reads.searchKbmaps parsed entries straight to hits, dropping a per-path de-duplication thatparseRipgrepOutputalready guarantees.🧪 Tests
recall.test.tspins the full ripgrep argument list, and asserts metacharacter escaping, disjunctive term combination, and alias expansion against the constructed pattern. All four failure branches (no-match exit, absent binary, other non-zero exits, unparseable output) run through a fake runner.parseRipgrepOutputtests over synthetic events, replacing four tests that proved the same point by matching real notes under date-patterned paths.search.test.tsdeclare which notes recall found, through a sharedbuildRecallStubhelper that resolves basenames to absolute paths and attributes each hit to its in-scope KB. Declaring the hit set is a stronger statement than the implicit matching it replaces, and it lets several assertions tighten from a lower bound to an exact count.rgshim that logs its arguments and exits non-zero: the full repository suite, the post-build smoke run, andnmr build && nmr check:strictall pass with an empty log.Closes #1094